Search

In this paper the problem of identifying a fuzzy model from noisy data is addressed. The piecewise-affine fuzzy model structure is used as non-linear prototype for a multi–input, single–output unknown system. The consequents of the fuzzy model are identified from noisy data which are collected from experiments on the real system. The identification procedure is formulated within the Frisch scheme, well established for linear systems, which is extended so that it applies to piecewise-affine, constrained models.

This paper investigates some of the many algebraic properties of the solution of the Frisch identification scheme applied to dynamic systems. These properties are related to the design of a robust selection criterion leading to a single model also when the assumptions of the scheme are not fulfilled.

The problem of determining linear models from data corrupted by noise has been revisited by several authors in recent years. With reference to the Frisch scheme, this work presents some geometrical aspects regarding the sets of solutions obtained in the identification procedure. An analysis is made on how these sets vary as the amount of noise on the data varies; the existence of particular solutions for data linked by a number of linear relations greater than one, is also shown.

The fuzzy model identification problem from noisy data is addressed. The piecewise linear fuzzy model structure is used as a nonlinear prototype for a multi-input, single-output unknown system. The consequent of the fuzzy model is identified using noisy data, e.g. collected from experiments on a real system. The identification procedure is formulated within the Frisch scheme, well established for linear systems, which has been modified and improved to be applied in fuzzy systems field.

The structural identification of linear muitivariabie systems can be performed on the basis of the predicted percentage of reconstruction errors (PPCRE) associated to a sequence of increasing-complexity models. This paper extends some previous results describing the properties of the PPCRE for canonical models whose parameters are estimated by means of the eigenvector method.

In this paper a class of input-output multistructural (overlapping) models is developed starting from the usual state-space ones. The algebraical link between state-space and input-output multistructural models is then investigated. The problems arising in the use of input-output multistructural models in the identification of multivariable systems are finally discussed and a procedure based on the use of these models proposed.

The identification of complex industrial processes on the only basis of input-output sequences obtained during normal operating conditions plays, at present, an important role-A distillation column, for instance, can work in many different conditions (feed flow, composition, temperature etc.) also if it has been designed wit|h reference to a standard situation. This requires the presence of a controller based on an accurate previsional model of the process. This paper describes the construction of a mathematical model for a ethane-ethylene distillation column by identification techniques .

This paper presents an identification method for errors-in-variables systems with input-output measurements affected by white and mutually correlated noises. The procedure, based on a frequency-domain approach, allows to uniquely determine both the characteristics of the noises affecting the data and the transfer function of the process under investigation. A numerical example is reported in order to illustrate the suggested technique and to verify its numerical implementation.

This paper concerns the identification of piecewise linear models from noisy data. The identification procedure is formulated within the Frisch scheme (1934), a technique well established for linear systems.